Anthropic caught scientists using Claude to further biological weapon research52m ago China AI Star Moonshot Eyes $2 Billion Annualized Sales in 20261h ago What's next for the AI safety debate1h ago AI researchers back an extinction warning as the Pentagon rejects it1h ago Congress gripped by AI panic after doomsday warnings1h ago How Meta tried, and failed, to grab its employees' data for AI1h ago Why So Many AI Researchers Think the Machines Could Kill Everyone1h ago Amazon adds Mandiant founder Kevin Mandia to its board2h ago Session Traces and Cost Controls Help Diagnose AI Agent Failures2h ago UK economy unexpectedly grows 0.4% in July boosted by AI2h ago The AI safety vibe shift2h ago Seed Capital closes a €130M fifth fund and expands from Denmark into the Nordics2h ago Bridgewater’s Jensen Says AI Will Kill People Before It’s Curbed2h ago 'Nothing new to announce' on possible chip collaboration with Samsung: OpenAI3h ago

Anthropic caught scientists using Claude to further biological weapon research52m ago China AI Star Moonshot Eyes $2 Billion Annualized Sales in 20261h ago What's next for the AI safety debate1h ago AI researchers back an extinction warning as the Pentagon rejects it1h ago Congress gripped by AI panic after doomsday warnings1h ago How Meta tried, and failed, to grab its employees' data for AI1h ago Why So Many AI Researchers Think the Machines Could Kill Everyone1h ago Amazon adds Mandiant founder Kevin Mandia to its board2h ago Session Traces and Cost Controls Help Diagnose AI Agent Failures2h ago UK economy unexpectedly grows 0.4% in July boosted by AI2h ago The AI safety vibe shift2h ago Seed Capital closes a €130M fifth fund and expands from Denmark into the Nordics2h ago Bridgewater’s Jensen Says AI Will Kill People Before It’s Curbed2h ago 'Nothing new to announce' on possible chip collaboration with Samsung: OpenAI3h ago

AI Research Society & Culture6 min reading time

Why So Many AI Researchers Think the Machines Could Kill Everyone

WiredPublished 1 hour ago

Several leading AI researchers, including former Google DeepMind and Anthropic employees, have expressed deep concerns about AI systems potentially achieving recursive self-improvement, which could lead to uncontrollable superintelligence and existential risks to humanity. Recent AI breakthroughs and security incidents have intensified these fears, with some experts estimating a significant probability of catastrophic outcomes within the next decade.

#ai-safety #anthropic #google-deepmind #jacob-coxon #nate-soares #recursive-intelligence #superintelligence-risk

More in AI Research

AI Research3 min read

Anthropic caught scientists using Claude to further biological weapon research

Covered by 4 sources13h ago

AI Research5 min read

AI researchers back an extinction warning as the Pentagon rejects it

Covered by 2 sources13h ago

AI Research4 min read

An Anthropic researcher just quit, saying OpenAI and Anthropic are 'gambling with our lives'

Covered by 13 sources2d ago